BIG Re-registration Selection Model

CIBG

A healthcare professional applying for BIG re-registration is not always required to support their application with supporting documents. A random and targeted sample is carried out. The BIG Register uses an algorithm for the targeted sample.

Last change on 23rd of June 2026, at 9:06 (CET) | Publication Standard 1.0

• Publication category: Impactful algorithms
• Impact assessment: DPIA
• Status: In use

General information

Theme

• Work
• Health and Healthcare

Begin date

2013-05

Contact information

[e-mail]

Link to publication website

https://www.bigregister.nl/service/privacy#36f7a06aab

Responsible use

Goal and impact

The aim is to select, from the targeted sample, primarily those applications that are highly likely to fail to meet the statutory requirements. Applicants who are likely to comply, on the other hand, should have their administrative burden reduced as much as possible. This approach seeks to balance the desire for effective protection of patients’ interests with the interests of healthcare professionals in keeping their administrative and financial burden to a minimum.

Considerations

The aim of the selection model’s categorisation is to reduce the administrative burden on healthcare providers and to operate more on the basis of trust rather than mistrust. In order to reduce the administrative burden on healthcare providers, it is necessary for them to be placed into categories. It is difficult to achieve a reduction in the administrative burden by any other means.

Human intervention

The selection model determines whether an application is selected for the submission of (objectively verifiable) supporting documents. These supporting documents are then assessed by at least one member of staff.

Risk management

1. Prevention of direct discrimination – when applications are assessed by the algorithm, no sensitive application characteristics (such as gender, age or foreign nationality) are shared on the basis of which the algorithm could make a direct distinction. As these are not provided as application characteristics, the algorithm cannot select on the basis of them. Furthermore, these characteristics are not included as relevant criteria in the selection process. This is to prevent direct discrimination on the basis of these characteristics;
2. Prevention of indirect discrimination – although the targeted selection process cannot directly select on the basis of sensitive file characteristics, it would theoretically be possible for a certain – seemingly innocuous – attribute might be included in the selection process, which would, in practice, result in certain groups of healthcare providers receiving high scores more frequently. This is prohibited if there is no justification for it. For this reason, the system is periodically monitored for indirect bias. This helps to determine whether, as a result of the selection process, certain groups of healthcare providers are selected relatively more often than others. Specifically, the following characteristics are examined: age, gender and whether or not the individual holds a foreign nationality;
3. In addition, we must prevent the algorithm from becoming self-reinforcing. For this reason, a random sample is taken alongside the targeted sample;
4. It is not apparent to the staff assessing the supporting documents whether the application in question has been included in the random or targeted selection. This is to prevent them from being unconsciously influenced by this information;
5. With regard to explainability, the following should be noted: different scores are assigned during the selection process. These scores are clear, specific and can be explained in relation to a specific case;

Legal basis

The BIG Act and the Decree on Periodic Registration under the BIG Act.

The explanatory memorandum states the following in this regard: Applications for periodic registration will not, as a rule, need to be supported by documentary evidence. The content of self-declarations will be checked on the basis of a random or targeted sample.

Links to legal bases

• The BIG Act: https://wetten.overheid.nl/BWBR0006251
• • Decision on periodic registration under the BIG Act: https://wetten.overheid.nl/BWBR0024841

Impact assessment

Data Protection Impact Assessment (DPIA)

Operations

Data

1. General information about the case/application
2. Personal information (this information is required to identify any indirect bias)
3. Information regarding hours worked
4. Information regarding the nature of the work carried out
5. Measures applicable abroad

Technical design

The system uses a points-based scoring system, based on a number of pre-defined target criteria, to determine which applications are included in the sample.

A score is assigned for certain characteristics. In addition, a maximum score has been set. If the sum of the scores for a particular application’s characteristics exceeds this maximum score, the application is always selected.

Applications that do score on the characteristics but do not exceed the maximum score may still be selected for the targeted sample. The targeted sample is a percentage that is set. The higher the total of the scores for the characteristics, the greater the chance that the application will be selected in the targeted sample.

External provider

Developed in-house